+ All Categories
Home > Documents > Michel Callon, Michel Foucault and the `` dispositif

Michel Callon, Michel Foucault and the `` dispositif

Date post: 02-Feb-2022
Upload: others
View: 28 times
Download: 0 times
Share this document with a friend
Embed Size (px)
of 12 /12
HAL Id: hal-00546736 https://hal.archives-ouvertes.fr/hal-00546736 Submitted on 20 Dec 2011 HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. Michel Callon, Michel Foucault and the “ dispositif ” Hervé Dumez, Alain Jeunemaitre To cite this version: Hervé Dumez, Alain Jeunemaitre. Michel Callon, Michel Foucault and the “ dispositif ”: When economics fails to be performative: A case study. Le Libellio d’AEGIS, Libellio d’AEGIS, 2010, 6 (4), pp.27-37. hal-00546736
Page 1: Michel Callon, Michel Foucault and the `` dispositif

HAL Id: hal-00546736https://hal.archives-ouvertes.fr/hal-00546736

Submitted on 20 Dec 2011

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

Michel Callon, Michel Foucault and the “ dispositif ”Hervé Dumez, Alain Jeunemaitre

To cite this version:Hervé Dumez, Alain Jeunemaitre. Michel Callon, Michel Foucault and the “ dispositif ”: Wheneconomics fails to be performative: A case study. Le Libellio d’AEGIS, Libellio d’AEGIS, 2010, 6 (4),pp.27-37. �hal-00546736�

Page 2: Michel Callon, Michel Foucault and the `` dispositif


Le Libellio d’ AEGIS

Vol. 6, n° 4 – Hiver 2010

pp. 27-37

Michel Callon, Michel Foucault and the « dispositif » When economics fails to be performative: A case study

Hervé Dumez & Alain Jeunemaître

CNRS / École Polytechnique

I n recent years, many important studies have focused on performativity or

performation of economics (among others, see Callon, Milo & Muniesa, 2007;

MacKenzie, Muniesa & Siu, 2007). Together with authors such as Yuval Millo,

Fabian Muniesa, Donald MacKenzie, or Lucia Siu, Michel Callon has played an

important role in this field. Most of these scholars have highlighted the way in which

economics performs real world. We will, in contrast, focus on the opposite. Drawing

on the case of air traffic management (ATM) in Europe, a case we have been working

on for about ten years, we will attempt to show how economics failed to perform this

industry. Our purpose is not to offer a kind of refutation or falsification of the

performativity theory from a Popperian stance. Obviously, this theory acknowledges

the fact that economics does not always perform the economy. Rather our aim is to

refine the theory, to pinpoint some paradoxes related to performativity or

performation, and to bring forward thinkable promising research perspectives.

We shall begin by examining what, in Michel Callon‟s perspective, constitutes and

does not constitute the performativity of economics. Then, we will have a look at

Michel Foucault‟s notion of « dispositif » (apparatus or device), to explore how

Callon relies on Foucault, but also how Foucault can still be useful when

approaching the notion of device. After that, we will present the selected case, air

traffic management (ATM) in Europe. Finally, we will discuss this case in connection

with the theoretical framework.

Michel Callon on performativity (or performation)

To understand what Michel Callon has in mind when using the notion of

performativity or performation, it is useful first to understand what it is not or what

it is opposed to.

First, performativity is opposed to embeddedness, to the idea that economy is

embedded in a social world that determines it. What seems interesting to Michel

Callon is not how the economy is embedded, but how it succeeds in gaining

autonomy from the social world. Michel Callon, here, is quite in line with Fernand

Braudel (1979/1992) when he showed how the economy tried to free itself, in a

difficult and slow process, from the social world, especially religion. Emphasis is

therefore placed on the mechanisms that allow the economy to become separate from

the social environment and, reciprocally, on how the social can thereafter be

economized: « the question is not: what do we call economic behavior, or what is the

Page 3: Michel Callon, Michel Foucault and the `` dispositif

Page 28

AEGIS le Libellio d’

economy, but how are behaviors, institutions, agencements, and rules of the game

economized? » (Callon, 2008, p. 22).

Principally since Walras (Dumez, 1985), distinctions are made between pure and

applied, and positive and normative economics. On the one hand, there would be the

abstract world of models, explicitly or implicitly normative, and on the other hand,

there would be the real world to which these models would be or should be applied.

The notion of performativity questions these distinctions.

Something similar can be said in connection with the concept of convention, or the

idea that agents must agree on rules and norms before acting, and in order to

interact. These rules, explicit or tacit, would be the necessary condition of

coordinated economic behaviors. According to Michel Callon, however, conventions

could be and must be forgotten. Performativity is something very different from

agreed upon rules.

Performativity can also be contrasted to self-fulfilling prophecies. If all agents

believe something will happen and act accordingly, it turns out the thing happens.

Michel Callon considers this mechanism to be simplistic, and not to correspond to the

way economics performs reality.

Finally, performativity must be distinguished from (neo) institutionalism. In order

to function, the economy is said to need institutions. Michel Callon thinks the notion

of institution is too static: it can explain how an economic state can reproduce itself,

but not how it can evolve. And institutions are stuck: they do not explain how they

can evolve themselves. They are a framework and, as such, seem unable to change

and let change happen. In Michel Callon‟s view, old as new institutionalisms lack

explanatory power.

Thus, to understand what Michel Callon means by performativity or performation,

we must first examine the large and impressive set of ideas from which this notion is

intended to break. These ideas include all the well-known theories regarding the

economic world, orthodox as much as heterodox, Nobel-prize winning or otherwise.

The second step consists of focusing on what Michel Callon concentrates when

speaking of performativity, namely the economic discourse and devices.

Regarding economic discourse, Michel Callon thinks it is important to go beyond the

distinction between the disciplinary, scientific, academic economic theory, and the

spontaneous economic theory by practitioners. To capture that, we introduced, when

analyzing price control practice in France between 1936 and 1986, the expression

« spontaneous (economic) theory » (Dumez & Jeunemaître, 1989). The academic

official theory condemns price control as an infringement of the free market

mechanism. The price controllers, some pure practitioners, other (like Taussig or

J. K. Galbraith) trained in academic economics, invented a sort of spontaneous

theory of the practice (« spontaneous philosophy » is an expression by Gramsci,

taken over by Althusser, 1967). When dealing with the performativity of economics,

Michel Callon (2006) speaks of « economics at large », which means academic

economic theory as economic “theory” made by practitioners, or spontaneous

economic theory. « Economics at large » is a domain that avoids the traditional

distinctions between pure and applied, positive and normative economics, and

subsumes the various types of economic discourse.

What Michel Callon has in mind when forging such a large and undetermined

domain, is to show that performativity relies on the creativity (models, algorithms,

spontaneous economic reasoning) of the agents – of all types of agents, practitioners

as well as academics. In that perspective, experimental economics appears as one of

Page 4: Michel Callon, Michel Foucault and the `` dispositif

Page 29

Volume 6, numéro 4

the most important development of recent years. During the initial period,

experimental economics was not very creative: the aim was to « test » models, to see

whether or not they matched agents‟ real behaviors. As economists are poor

Popperians, and behave themselves rather as scientists in Thomas Kuhn‟s

description, i.e. they are not very interested in falsifying their theories, experimental

economics was for them mostly a matter of indifference. Things changed when

experimental economics no longer limited itself to testing models, but sought to

invent new forms of markets. Michel Callon believes that, through the experimental

approach, economics is enlarging the gap between itself and other social sciences:

« L’expérimentation à la fois comme creuset où s’élabore la théorie et comme cadre qui

permet à cette théorie d’avoir des effets, est devenue selon moi un enjeu

majeur » [« experimentation is at the same time the crucible in which theory is

elaborated, and the framework allowing this theory to have effects; today,

experimentation has become a major challenge »] (Callon, 2006, pp. 26-27). Fabian

Muniesa and Michel Callon speak of « experimental performation ».

Performativity can be thus rooted in theory or practice. In this process, argues

Michel Callon, devices are essential. In an interview, Michel Callon (2008) explains

that, since The laws of the markets (1998), he may have changed his mind on

performativity, which became performation or coperformation, but not regarding

devices. These are the central, invariant, elements of the approach (Callon, Milo &

Muniesa, 2007; MacKenzie, Muniesa & Siu, 2007). There is a paradox here. Strictly

speaking, performativity designates a discourse that is at the same time an act that

changes the world. The change in the world coincides with the uttering. The first

historical statement of performativity, the one by Varro in his De lingua latina, is

clear-cut: « Spondere est dicere spondeo » (« To promise is to say: I promise »). In

Michel Callon‟s perspective, the performativity or performation of economics

definitely does not work in that way. Economics performs the real world if and only

if socio-technical devices exist that make the performation possible. If devices are

few and limited, the performation of economics is rare and limited (Callon, 2006, p.

26). Theory becomes true if devices let it become true. In the perspective of the Actor

Network Theory (ANT), devices as non-human actors act, and in a sense even act

more truly, than human actors. Michel Callon underlines two additional points.

Upstream, devices are rooted in algorithms. These algorithms may be anchored in

academic theory, and be presented and discussed in economic journals, or in practice

and spontaneous theory. Upstream again, in order for economic devices to work,

goods must have been created as abstract or concrete objects that are both valuable

and exchangeable. Downstream, Michel Callon insists on the calculative skills with

which the agents must be equipped. But the analytical core of the approach is the

device itself. A simple algorithm, the creation of valuable goods, the agents‟

calculative equipment not by themselves sufficient to make economics performative.

For instance, the agents‟ calculative equipment is created by the devices, and not the

other way around, even if there is always at the same time framing and overflowing

(which makes the difference with institutionalism). Since Michel Callon refers to

Michel Foucault when speaking of devices and performativity, it might be useful to

turn to the latter to understand what devices are in the former‟s perspective.

Michel Foucault and the « dispositif » (device, or apparatus)

Michel Foucault uses the French word « dispositif », which is translated as

« apparatus » (Agamben) or device (Callon), and sometimes let in French as English

translations are not completely satisfying. Actually, the word « dispositif » comes

Page 5: Michel Callon, Michel Foucault and the `` dispositif

Page 30

AEGIS le Libellio d’

from very far away in the past. Agamben (2009) has tried to draw its genealogical

tree. In French, « dispositif » originally means the final part of a sentence, where

reasons of the decision reached by the judge are explained in an ordered way. Then it

means the way a general intentionally disposes his forces to get the maximum

defensive or offensive impact. But Foucault takes the word from his master in

philosophy, Jean Hyppolite. Hyppolite wrote on the young Hegel‟s Die Positivität

der Christlichen Religion, where Hegel drew a distinction between natural religion, a

direct contact with divinity, and the positive religions made of prescriptions,

discourses and rituals that constrain religious behaviors. In this context,

« dispositif » comes from the Latin dispositio, which translates the Greek oekonomia

in a religious sense. The word has been used, especially by Ireneas of Lyon, to mean

the way God intentionally organizes the salvation of humanity – what is called the

economy of salvation (Fantino, 1994).

As for Foucault, the crucial reference in this connection is an interview he gave to

the psychoanalytic French journal Ornicar in 1977:

[By « dispositif », I mean:] firstly, a thoroughly heterogeneous ensemble

consisting of discourses, institutions, architectural forms, regulatory

decisions, laws, administrative measures, scientific statements, philosophical

and moral propositions – in short, the said as much as the unsaid. Such are

the elements of that apparatus. The apparatus itself is the system of

relations that can be established between these elements. Secondly, what I

am trying to identify in this apparatus is precisely the nature of the

connection that can exist between these heterogeneous elements. Thus, a

particular discourse can figure at one time as the program of an institution,

and at another it can function as a means of justifying or masking a practice

which itself remains silent, or as a secondary re-interpretation of this

practice, opening out for it a new field of rationality. In short, between these

elements, discursive or non-discursive, there is a sort of interplay of shifts of

position and modifications of function, which can also vary very widely.

Thirdly, I understand by the term „apparatus‟ a sort of – shall we say –

formation which has as its major function at a given historical moment that

of responding to an urgent need. The apparatus thus has a dominant

strategic function […] there is a first moment, which is the prevalent

influence of a strategic objective. Next, the apparatus as such is constructed

and enabled to continue in existence insofar as it is the site of a double

process. On the one hand, there is a process of functional over-determination

[…] on the other hand, there is a perpetual process of strategic elaboration.

(Foucault, 1980, pp. 194-195)

Why is this text interesting in our perspective? First, because it shows the direct

connection between Michel Callon‟s and Michel Foucault‟s analysis of devices. A

device (or apparatus, i.e. a « dispositif ») has a hybrid nature. It is made of discourses

and non-discourses (« du dit et du non dit »). It is the relationships between these

heterogeneous elements that really count, and they can be complex. A discourse can

be the programme of an institution, or, on the contrary, be disconnected from

practices and conceal them (the practices being themselves silent, « muettes »). But it

can also be an a posteriori interpretation of practices that developed ahead of it. One

can see in Michel Callon‟s work the same complex relationships affecting discourses

and non-discourses, the idea that devices structure behaviors and therefore act, the

notion that devices are made of knowledge and, at the same time, produce


The second reason for paying attention to Michel Foucault here is that one element

present in his text seems to have (at least partially) disappeared in Michel Callon‟s

Page 6: Michel Callon, Michel Foucault and the `` dispositif

Page 31

Volume 6, numéro 4

work, i.e. the particular dynamics of devices as structured in two phases. The

creation of devices is marked by an « urgent need » (as Michel Foucault puts it).

Then, during the second phase, new, unanticipated functions, strategies, and

processes emerge and contribute to stabilize and entrench the device (if it does not

rapidly disappear). So Michel Foucault‟s vision is an invitation to focus on two

distinct moments: the appearance of the device, and its stabilization, a strategy that

uncovers continuities and discontinuities.

Let us now analyze a case study to discuss Michel Foucault‟s and Michel Callon‟s

ideas on performation and devices.

The case of the European Air Traffic Management

The selected case is an industry that, at its very beginning, was not conceived as a

market and was kept off from the usual economic models and tools (profit,

competition), and is now exposed in Europe to performation or performativity of

economics: the Air Traffic Management (ATM), which consists mainly in Air Traffic

Control, but also in flow management1.

The control of the movements in the

sky goes back to the Franco-Prussian

war of 1870-1871, when balloons were

used by both armies: Did states have

the right to control the sky above their

territory? The question became more

topical at the time of the first airplane

race in 1909. Blériot, the winner, flew

accross the English Channel, taking off

in France to land in the United Kingdom. An international conference was set up in

Paris the following year, but no agreement was reached. Two principles of law

oppose each other in this context. The first one is the sovereignty of nation states.

Could a foreign plane be allowed to fly over Paris and land somewhere in France

without the French State having been informed and having authorized it? The

second one was formulated by Grotius in 1609 in his Mare liberum, and establishes

that the seas – here, the skies – cannot be the property of a state and must remain

free of access. The debate raged for thirty years until 1944, when the Chicago

Convention was signed by more than fifty states. The Convention stipulates that a

plane can freely fly over a country, provided it does not land, or lands to refuel

without embarking or disembarking any passenger or good. But each country is

entitled to define routes and to control the planes flying over its territory. Countries,

however, are not allowed to set up a toll. With the development of jets, security

became a major problem.

A propeller generally has enough time left to avoid collision with another propeller

when both can see each other. But when a jet notices another jet, it is usually too

late, and the collision is inevitable. It therefore becomes necessary to define routes

and corridors, to create radar coverage in order to determine where each aircraft is

located, to organize phone calls so that controllers can give instructions to the pilots.

This costs a lot. At the beginning of the 1970s, it was admitted that the aircraft

crossing the sky of a country must pay for the development of these technically

sophisticated and heavy systems. It remained clear that this involved only cost

recovery, and not a toll to give access to the sky, which remained free for all. In the

which US, the air traffic management was provided by a public administration (the

Federal Aviation Authority), and a tax collected on each ticket (and indicated as

1. For a more detailed

presentation, see Beyer,

2008; Brooker, 2003; Button

& McDougall, 2006; Dumez

& Jeunemaître, 2001; 2010.

le Blériot XI

qui a traversé la Manche

le 25 juillet 1909

Page 7: Michel Callon, Michel Foucault and the `` dispositif

Page 32

AEGIS le Libellio d’

such on the ticket itself). In Europe, each aircraft pays route charges to every

country it has crossed the sky. In the 1970s, the air carrier market was regulated,

competition was limited, and profits were substantial. To pass the cost of air traffic

management on to the passengers was not a big issue.

We have presented (very briefly) the case. It is striking that the ATM industry was

maintained off the usual economic rules of competition and profit. Two questions

then arise: can the ATM be analyzed as a device, and, if yes, in what sense? If it can

be analyzed as a device, what is it the performation of?

ATM both presents a unity and is heterogeneous. The unity comes from an intention

or intentionality: making flight as free and secure as possible. The heterogeneity

comes from an intertwining of rules (both legal and technical) and technical systems

(radars, telecoms, screens of control, huge software that compute flight plans data

and helps controllers in their task), i.e. from an intertwining of discourses and non-

discourses. This complex heterogeneous system governs the behavior of pilots, air

carriers, controllers, and states. It relies on legal and technical knowledge and know-

how, and it produces knowledge, as any control system does. Insofar as it

encompasses algorithms, there are good reasons for considering it as a device in

Foucault‟s and Callon‟s sense. This applies, for instance, to the determination of the

en-route charges aimed at recovering costs. The algorithm takes into account the

distance flown in the sky of the country crossed over, the square root of take-off

weight of the aircraft, and a service unit defined by each country on the basis of the

costs of the national ATM system and of the number of controlled flights per year.

Another algorithm manages the disequilibria between the flight plans asked for by

the air carriers and the capacity of control in real time.

If the answer to the first question is that there are indeed good reasons to see the

ATM as a device in Foucault‟s and Callon‟s sense, then we can turn to the second

one: what is this device the perfomation of? Analysis tends to indicate it is the

performation of law with the help of engineering. Law, a subtle and complex

combination aimed at managing liberty of flight and national sovereignty

simultaneously, thus relied on engineering to establish the device.

Now comes a third question: what place was given to economics in that device?

Economics was intentionally contained and its place reduced to a minimum. Profit

was eliminated; the only implemented economic principle was that the en-route

charges must be cost-related. But the cost-relation is weak: each country announces

its costs and recovers them. There is no in-depth control of these costs. However, an

attentive look demonstrates that economics is not completely absent. To a controller,

any aircraft, whatever its type, is a point on a screen and must be handled so as to

avoid potential collisions. The aircraft is controlled as long as it stays within a

country‟s sky. Costs are related to control duration, and, as an approximation that is

easier to calculate, to the distance flown in the national sky. In Europe, however,

another factor is taken into account: the takeoff weight of the aircraft. While the

task of control is exactly the same for both large and small airplanes, the former pay

more, though only in relation to the square root of their weight. Big machines thus

cross-subsidize small ones, but not too much. The algorithm organizes cross-

subsidization. More generally, the device as a whole, supposed to perform law and

engineering but not economics, entails nonetheless cross-subsidization. We will

return to that crucial point. The answer to the third question is mixed. Economics

was not present in the device when it was created, and then it penetrated it in a very

limited way.

Page 8: Michel Callon, Michel Foucault and the `` dispositif

Page 33

Volume 6, numéro 4

Things changed in the 1980s when air transportation was deregulated. Competition

became fierce with the apparition of low cost companies. This had two main

consequences: operational margins, which were comfortable until then, decreased

dramatically and traffic exploded. As regards ATM, it appeared at the same time

very costly (eroding the operational margins of the air carriers) and inefficient

(delays grew, peaking in 1988-89). This raised the issue of performance. Economics

was called for at two different levels: the transformation of the whole industry into a

market, and the introduction of limited economic mechanisms. In the 1990s it was

discussed to merge all the different national upper airspaces into one European

unified upper airspace, and to organize auction procedures for the entire block or a

few great partial blocks. This assumed that part of the airspace was transformed

from an object of national sovereignty into a valuable, exchangeable, good.

Competition was set up – not in the market, but for the market – through an auction

mechanism between the different national service providers, individually or allied.

Technically, specialists agreed that it would be possible to control the entire

European upper airspace with four of five centers (in the current situation, each

country has at least one center, large countries such as France, Germany or the

United Kingdom have several). The creation of a market lay within the framework

of a general development where the deregulation of air transportation and the

privatization of airports and control towers had created real markets. How could the

European ATM have lasted as a legal-technical device, like an isolated islet on an

ocean of markets?

However, the insertion of the ATM in an economy of competition and profit

stumbled against the sovereignty of nation states. In particular, states put military

issues in the forefront to keep in place their national service providers of air traffic

control. It was an excuse more than anything else, since the European military are

aware that an efficient air defense is impossible in the framework of one single

country: Paris is only two hundred kilometers off the boundary with Belgium. At a

time of jet fighters and crusader missiles, the air defense of Paris takes place far away

from French boundaries. That is why European Air Forces strongly cooperate within

NATO. But of course, it is also true that they must be able to take over the national

sky in case of a major crisis. This can be guaranteed, even if the upper airspace is

normally managed on a European basis.

The crucial issue was not really there. Any transformation of the European ATM

industry into a market would have deeply changed the multiple cross-subsidization

mechanisms established by the legal-technical device. As said before large aircraft

pay for smaller ones, and big international air carriers for regional companies.

Regional companies could therefore disappear with the introduction of market

mechanisms. Besides, the separation between upper and lower airspace emphasizes

the fact that upper airspace is far less costly to control than the lower. In the upper

airspace, aircraft remain at the same altitude. In contrast, when approaching the

ground, they change altitude and direction. The control task is far more complex and

costly. When upper and lower airspace are not separated, high-altitude flights pay

for flights that leave or approach airports. US companies that fly over Europe pay

for European carriers that use airports in the continent. If, however, upper and lower

airspaces were separated, US companies would end up paying less, and European

ones more. The development of a market would make visible the existing hidden

cross-subsidization; thus, European air carriers, which would be inclined to favor the

development of a more efficient market, actually block it.

Page 9: Michel Callon, Michel Foucault and the `` dispositif

Page 34

AEGIS le Libellio d’

At the same time, the legal-technical device has shown a true capacity of innovation.

When delays reached a maximum, an organization, the Central Flow Management

Unit, was set up to manage in real time the disequilibria between the demand from

the air carriers and the capacities offered by the European service providers of air

traffic control. The disequilibria could have been interpreted as a classic economic

issue, as an imbalance between supply and demand to be dealt with by such

economic tools as congestion pricing. The economic approach was discussed, but not

adopted. Rather, a management algorithm was chosen, which follows the principle

of first-come, first-served. It processes flight plans one after the other and, if demand

exceeds supply (the capacity of control), it stops aircraft on the ground. Before the

implementation of that algorithm, aircraft took off and were delayed in flight. This

costed a lot in fuel. The new device has therefore positive economic consequences.

Nevertheless, it is not a performation of theoretical economic ideas or of a

spontaneous economic reasoning, but rather an engineering device with a law

principle: treat every aircraft, which means every company, big or small, on an equal

footing. In the same perspective, the European Commission, with its plan for the

industry (The Single European Sky), created another legal-technical device, the

Functional Airspace Blocks. It is inspired by the idea that considerable technical

efficiency gains could come from organizing control according to airspace blocks

larger than the national airspaces. The FABs were created by a European rule (law)

with a technical basis (they must be « functional »); they therefore constitute a legal-

technical device. The economic dimension was left backstage.

At the very beginning, FABs were seen as concerning only upper airspace. But the

upper/lower separation was later abandoned to avoid cross-subsidization. Another

example is given by en-route charges. They remain collected by the states for the

flown distance in their own sky. Let us suppose France and Switzlerland decide to

establish a FAB, putting together part of their airspace. If the aim is a better

functional performance, routes will probably be modified, and become more rational

than the previous ones, which took into account the boundary between France and

Switzerland (the new route will be shorter in the French sky, for example, and longer

in the Swiss sky). But every change of a route will create a loss for one of the

countries, and a mechanical gain for the other (here a loss for France and a gain for

Switzerland). Technical restructuration is therefore impeded by its economic


In conclusion, performation by economics failed to replace performation by law, with

the helping hand of engineering to shape the device. This failure operates at the

general level of attempting to transform the industry into a market. Economic

elements were nonetheless introduced into the legal-technical device. Some public

service providers have been transformed into economic agents through privatization.

In the United Kingdom, for example, a licence has been given for thirty years to an

alliance of airline companies led by British Airways through an auction mechanism.

The framework nonetheless remains the entire national sky, not separated into an

upper and a lower airspace. Economic agents have been set up, but without the

creation of a European market device. Economics stepped in with the proposal for

substitutes for market mechanisms, something on which academic economists have

been working since the 1980s. Such mechanisms mainly consist in regulatory policies.

The European Commission imposed on each country a separation between regulation

and service provision, and established a regulator. As a result, for many countries,

the process included the creation of a privatized economic agent at the level of

service provision, and of an independent economic regulator at the public

administration level, without, however, leading to the creation of a market. At the

Page 10: Michel Callon, Michel Foucault and the `` dispositif

Page 35

Volume 6, numéro 4

same time, a « sunshine » regulatory authority, the Performance Review

Commission, was set up at the European level, disposing of no real power except the

one to publish reports on performance issues (McCraw, 1984). Thus, some economic

elements were incorporated into a legal-technical device.


Obviously, it would be risky to generalize from this case, and simplistic to think the

case could falsify or refute the performativity approach to economics. Interpretation

is far more complex. We will therefore insist on further potential research paths.

1. In Michel Callon‟s approach to performation or performativity, two

apparently contradictory tendencies seem to coexist. On the one hand, Michel

Callon states that, like other social sciences, economics is performative (see the

debate in Callon, 2006). On the other hand, when analyzing performativity,

Michel Callon focuses on economics in a particular manner, and, as we

discussed above, emphasizes some particular features, such as the current

trend in experimentation. Our case suggests that the two positions are not

really contradictory. One could think that two social sciences such as law and

economics have a particular status in terms of performativity (the idea that

law is different from other social sciences has been developed by Günther

Teubner from the perspective of autopoiesis; see for example Teubner, 1987).

Our case suggests that the particular status of law and economics derives from

the fact that both can at the same time define great principles (liberty,

equality and due process for law, performance for economics), and mobilize

techniques to set up devices which will operate the performative process. Law

and economics sometimes compete with each other, and that brings about

situations in which one dominates over the other, and in which the dominated

formulates strong critiques vis-à-vis the dominant. In our case, law dominates,

and economics tries to step in from the performance perspective. If our

analysis is right, the opposite situation might also be found. The question

remains open of whether other social sciences could compete with law and

economics in defining great principles to evaluate empirical situations and

mobilizing techniques to set up devices.

2. When analyzing the competitive process of performativity, it seems interesting

to pursue Foucault‟s dynamic approach, which separates moments of urgency,

especially during the creation of devices, from periods of routine. When the

ATM device was established, the object of urgency was not performance, but

the litigation between national sovereignty and liberty of flight. A legal-

technical device was adopted instead of an economic one. Only long after that,

the performance issue surfaced, in a situation characterized by a legal-

technical performation. The change towards an economization of the industry

ran against three main elements. First, inertia or path-dependence. Williamson

(1999) has highlighted that establishing a market in a different institutional

frame must be analyzed with the “remediableness” criterion in mind. Indeed,

it is always difficult to compare an existing institutional framework, with its

positive and negative, but long-operating elements, with an ideal alternative.

Status quo is not an option like others, since there is a premium on existing

devices (Beach, 1990). The second element that opposes the movement against

economization is the legal-technical device‟s capacity to innovate. As seen

above, the device was able to invent non economic algorithms to manage the

discrepancy between supply and demand of air traffic control or to establish a

Page 11: Michel Callon, Michel Foucault and the `` dispositif

Page 36

AEGIS le Libellio d’

legal-technical as complex object as the FABs. Third, the existence of cross-

subsidization. Agents that seemed to favor the industry‟s economization, e.g.

air carriers, turned out in the end to be very conservative, and sought to

maintain the legal-technical device to preserve the hidden financial flows of

cross-subsidization. There thus seems to be « moments of performation », and

these moments could be of different kinds.

3. The case highlights the difficulty of identifying mechanisms through which a

law-technical device could give way to an economic device. In examining the

performation of economics, Michel Callon has identified three elements:

algorithms, devices that provide agents with calculative equipment, and

valuable, exchangeable goods. For these elements to function in our case,

public administrations would have to become calculative agencies, the

national airspace, which is an object of sovereignty, would have to be

considered a valuable good, and market devices and algorithms would have to

be established. But is it possible to set up all of this at the same time, and thus

at once replace a legal-technical device that evolved progressively over a

century? Which mechanisms could enable such a transformation, i.e. an

economic performation of the industry? The most favored one has been

privatization. A market was supposed to emerge if agents were to transform

themselves into calculative agencies. In 1998, privatized service providers

established a trade association, CANSO to promote economic behavior. As we

explained, however, two obstacles hindered the development of a market: the

existence of a competitor device, and the fact that the airspace did not become

a valuable good. In Michel Callon‟s view, the device is the crucial factor.

Devices act, more than individual agents do. But who are the agents that push

and orient the creation of a device, and with what purpose? We here turn

again to Foucault: devices are established at particular moments marked by a

sense of urgency, and with an intent linked to this urgency. The initial aim can

disappear and be replaced by another, as well as by multiple functions that

make the device last. But are there other types of mechanisms to create

devices? This question seems to remain open.


Agamben Giorgio (2009) ‘What is an apparatus?’ and other essays, Stanford, Stanford

University Press.

Althusser Louis (1967) Philosophie et philosophie spontanée des savants, Paris, Maspéro.

[(1990) Philosophy & the Spontaneous Philosophy of the Scientists, and other essays, Gregory

Elliot (ed), London, Verso.]

Beach Lee Roy (1990) Image theory : decision making in personal and organizational contexts,

Chichester/New York, John Wiley.

Beyer Antoine (2008) “Les frontières du „Ciel unique européen‟. Enjeux techniques et

territoriaux du contrôle aérien en Europe”, Flux, n° 71, janvier-mars, pp. 8-23.

Braudel Fernand (1979) Civilisation matérielle, économie et capitalisme. XVe-XVIIIe siècles.

3 vol, Paris, Armand Colin. [1992, Civilization and Capitalism. 15th-18th, Berkeley (LA)

London, University of California Press.]

Brooker Peter (2003) “Single Sky and Free Market”, Institute of Economic Affairs Review,

June, pp. 45-51.

Button Kenneth & McDougall Glen (2006) “Institutional and structure changes in air

navigation service-providing organizations”, Journal of Air Transport Management,

vol. 12, pp. 236-252.

Page 12: Michel Callon, Michel Foucault and the `` dispositif

Page 37

Volume 6, numéro 4

Callon Michel [ed.] (1998) The Laws of Markets. Oxford, The Sociological Review/Basil


Callon Michel (2006) “La performativité de l‟économie”, Le Libellio d’Aegis, n° 3, pp. 21-28.

Callon, Michel, Millo Yuval & Muniesa Fabian [Eds.] (2007) Market devices, Oxford,


Callon Michel (2008) “Elaborating the notion of performativity”, Le Libellio d’Aegis, vol. 5,

n° 1, pp. 18-29.

Dumez Hervé (1985) L'économiste, la Science et le Pouvoir. Le cas Walras, Paris, PUF.

Dumez Hervé (2009) “Qu‟est-ce qu‟un dispositif ? Agamben, Foucault et Irénée de Lyon dans

leurs rapports avec la gestion”, Le Libellio d’Aegis, vol. 5, n° 3, automne, pp. 34-39.

Dumez Hervé & Jeunemaître Alain (1989) Diriger l'économie : l'État et les prix en France 1936-

1986, Préface de Jean Denizet, Paris, l’Harmattan collection “Logiques économiques”.

Dumez Hervé & Jeunemaître Alain (2001) “Improving Air Traffic Services in Europe. The

Economic Regulation Perspective” in Henry Claude, Jeunemaître Alain, & Matheu

Michel, Regulation of Network Utilities: The European Experience, Oxford, Oxford

University Press, pp. 290-311.

Dumez Hervé & Jeunemaître Alain (2010) “The managament of organizational boundaries. A

case study”, M@n@gement, vol. 13, n° 3, pp. 151-171.

Fantino Jacques (1994) La théologie d’Irénée. Lecture des Écritures en réponse à l’exégèse

gnostique, Paris, Éditions du Cerf.

Foucault Michel (1980) [edited by Gordon Colin] Power/Knowledge: Selected Interviews and

Other Writings, 1972-1977, New York, Harvester Press.

McCraw Thomas K. (1984) Prophets of Regulation, Cambridge (Mass), Harvard University


MacKenzie Donald, Muniesa Fabian & Siu Lucia (2007) Do Economists Make Markets? On

the Performativity of Economics, Princeton, Princeton University Press.

Muniesa Fabian & Callon Michel (2009) “La performativité des sciences économiques” in

Steiner Philippe & Vatin François [ed.] (2009) Traité de sociologie économique, Paris, PUF

Quadrige, pp. 303-338.

Teubner Gunther [Ed.] (1987) Autopoietic law. A new approach to law and society, Berlin, de


Williamson Oliver E. (1999) “Public and private bureaucracies: a transaction cost economics

perspectives”, The Journal of Law, Economics, and Organization, vol. 15, n° 1,

pp. 306-342